Deep Learning Framework for Groupwise Medical Image Registration

PhD Thesis Proposal Defence


Title: "Deep Learning Framework for Groupwise Medical Image Registration"

by

Miss Ziyi HE


Abstract:

Groupwise image registration (GIR) is a fundamental task that facilitates the
simultaneous deformation of subjects towards a specified or implicit group
center. Traditional methods usually suffer from substantial optimization
runtime, which limits their application in clinical tasks. With the advent of
deep learning methods, some related research works have emerged, but they
purely utilized the network as feature extractions, focusing more on
optimization instead of prediction.

This thesis intends to propose robust and efficient deep learning-based
frameworks for medical image groupwise registration. We start with an
unsupervised end-to-end groupwise registration framework with the multi-step
updating mechanism to align the group subjects into the latent group center
without explicitly constructing the template image. After that, we propose a
template synthesis method based on the generative adversarial network and an
auxiliary segmentation module to generate high-quality template images. To
extend the framework from the setting of fixed group size to arbitrary group
size, we present SETGen to deploy a Siamese variational autoencoder for
encoding pairs of inputs and generating the template image of the minimum group
unit through latent vectors' arithmetic. The method exhibits promising
flexibility and efficiency. In order to further improve the adaptability and
performance of SETGen, we propose TAG to integrate test-time training to SETGen
to deal with target groups of multiple resolutions. Experiments illustrate the
method outperforms state-of-the-art benchmarks as well as maintains the
robustness under various scenarios.


Date:                   Monday, 20 November 2023

Time:                   2:00pm - 4:00pm

Venue:                  Room 5562
                        lifts 27/28

Committee Members:      Prof. Albert Chung (Supervisor)
                        Prof. Chi-Keung Tang (Supervisor)
                        Prof. Long Quan (Chairperson)
                        Prof. Weichuan Yu (ECE)


**** ALL are Welcome ****